info_dict['pearsonr'] = pearsonr(preds.flatten(), Y_test.flatten())
nlpd = model.log_predictive_density(X_test, Y_test)
info_dict['nlpd'] = np.mean(nlpd)
# Get parameters
if args.model == 'ridge':
info_dict['coefs'] = list(model.coef_)
info_dict['intercept'] = model.intercept_
info_dict['regularization'] = model.alpha_
elif args.model == 'svr':
info_dict['regularization'] = model.best_params_['C']
info_dict['epsilon'] = model.best_params_['epsilon']
info_dict['gamma'] = model.best_params_['gamma']
else:
print model
param_names = model.parameter_names()
for p_name in param_names:
if p_name == 'warp_tanh.psi':
info_dict[p_name] = list(
[list(pars) for pars in model[p_name]])
else:
try:
info_dict[p_name] = float(model[p_name])
except TypeError: #ARD
info_dict[p_name] = list(model[p_name])
info_dict['log_likelihood'] = float(model.log_likelihood())
# elif args.model == 'rbf':
# info_dict['variance'] = float(model['rbf.variance'])
# info_dict['lengthscale'] = list(model['rbf.lengthscale'])
# info_dict['noise'] = float(model['Gaussian_noise.variance'])
# info_dict['log_likelihood'] = float(model.log_likelihood())
Y_metadata=noise_dict))
info_dict[emo] = emo_dict
preds_list.append(preds.flatten())
vars_list.append(vars.flatten())
# Get parameters
if args.model == 'ridge':
info_dict['coefs'] = list(model.coef_)
info_dict['intercept'] = model.intercept_
info_dict['regularization'] = model.alpha_
elif args.model == 'svr':
info_dict['regularization'] = model.best_params_['C']
info_dict['epsilon'] = model.best_params_['epsilon']
info_dict['gamma'] = model.best_params_['gamma']
else:
param_names = model.parameter_names()
for p_name in param_names:
if p_name == 'ICM.B.W':
info_dict[p_name] = list([list(pars) for pars in model[p_name]])
else:
try:
info_dict[p_name] = float(model[p_name])
except TypeError: #ARD
info_dict[p_name] = list(model[p_name])
info_dict['log_likelihood'] = float(model.log_likelihood())
# elif args.model == 'rbf':
# info_dict['variance'] = float(model['ICM.rbf.variance'])
# info_dict['lengthscale'] = list(model['ICM.rbf.lengthscale'])
# info_dict['noise'] = list([float(noise) for noise in model['mixed_noise.*']])
# info_dict['log_likelihood'] = float(model.log_likelihood())
# elif args.model == 'mat32':
